Managerial Economics in a Global Economy, 5th Edition by Dominick Salvatore

Chapter 5 Demand Forecasting

Prepared by Robert F. Brooker, Ph.D.

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Slide 1

Aims of Economic Forecasting

Reduce uncertainty Planning for long term growth
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 2

Kinds of Forecasts
Quantitative Nave Simple Short run
Barometric
Prepared by Robert F. Brooker, Ph.D.

Qualitative Sophisticated Complex Long run

Input-Output
Slide 3

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Factors Determining the Choice of Forecasting Method

Cost of Preparing the Forecasts and the benefit that results from it. Lead time in decision making Short run / Long run Forecast Level of accuracy required Quality and availability of data Level of complexity of relationships
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 4

Qualitative Forecasts
Used to make short term forecasts Unavailability of quantitative data Supplementing quantitative forecasts

Survey Techniques Soliciting a Foreign Opinion Polls Perspective

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 5

1. Survey Techniques
Rationale: Decisions are made well in advance of actual expenditures. Surveys of Economic intensions are being published by various bodies with fairly good predictability US firms spend more than $ 1 billion each year for surveys of consumers However consumers are now showing signs of sensitivity engaging in surveys regarding issues of privacy and time consumption.
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 6

2. Opinion Polls
Through Polling experts either with in the firm or from outside Executive polling: from inside / outside top management having good understanding Sales Force polling: as they are the people closest to the market Consumer Intentions polling: Polling a sample of potential buyers on their purchasing intensions
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 7

3. Soliciting a Foreign Response

Especially for firms having international concerns either from supply or from demand side Getting perspective from distinguished foreign dignitaries and business people.

Prepared by Robert F. Brooker, Ph.D.

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Slide 8

Time series Analysis

Nave Forecasting :
Underlying assumption is that time series will continue to move as in the past.

Prepared by Robert F. Brooker, Ph.D.

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Slide 9

Reasons for Fluctuations in Time Series Data

Secular Trend
Long-Run Increase or Decrease in Data

Cyclical Fluctuations
Long-Run Cycles of Expansion and Contraction

Seasonal Variation
Regularly Occurring Fluctuations

Irregular or Random Influences

Wars, natural disasters, strikes

**Total variation is the result of all 4 kinds of variations**

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 10

Prepared by Robert F. Brooker, Ph.D.

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Slide 11

Trend Projection
Linear Trend:

St = S0 + b t
S0 b

Estimated value in the base year Growth per time period Estimated value in the base year

St = 11.90 + 0.394 (t) St = 11.90 + 0.394 (17) = 18.60

Value of 1st Quarter 2004
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 12

Trend Projections for Quarterly Data

St = 11.90 + 0.394 (18) = 18.99 2nd Q St = 11.90 + 0.394 (19) = 19.39 3rd Q St = 11.90 + 0.394 (20) = 19.78 4th Q However the constant rate of change over time might not results close to reality. Then the constant percentage change over time might be a better idea.
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 13

Trend Analysis
Constant %age Growth Constant Growth Rate St = S0 (1 + g)t g = Constant Growth rate
In order to estimate g we would need to transform time series data in to their natural logs and then running regression on this transformed data

Estimation of Growth Rate lnSt = lnS0 + t ln(1 + g) (Linearized form) Because regression demands linearity. lnSt = 2.49 + 0.026t (antilog of ln(1+g) is 1.026) To make it interpretable we take antilog St = 12.06(1.026) t (antilog of 2.49 is 12.06) St = 12.06(1.026) 17= 18.66 (as against 18.60)
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 14

Seasonal Variation
If the data shows consistent variations corresponding to long run trend values. For example values of certain time periods (quarters) is below the trend and some are above the trend consistently, we identify the seasonal variation and deal with it either through Ratio to Trend Method or with dummy variables.
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 15

Ratio to Trend Method

We simply find the average ratio (of actual and forecasted values) in each quarter and the multiply the forecasted trend value by the average ratio.

Prepared by Robert F. Brooker, Ph.D.

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Slide 16

Seasonal Variation
Ratio to Trend Method Actual Ratio = Trend Forecast Seasonal Average of Ratios for = Adjustment Each Seasonal Period Adjusted Forecast
Prepared by Robert F. Brooker, Ph.D.

Trend = Forecast

Seasonal Adjustment
Slide 17

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Seasonal Variation
Ratio to Trend Method: Example Calculation for Quarter 1
Trend Forecast for 1996.1 = 11.90 + (0.394)(17) = 18.60 Seasonally Adjusted Forecast for 1996.1 = (18.60)(0.8869) = 16.50

Year 1992.1 1993.1 1994.1 1995.1

Prepared by Robert F. Brooker, Ph.D.

Trend Forecast Actual 12.29 11.00 13.87 12.00 15.45 14.00 17.02 15.00 Seasonal Adjustment =

Ratio 0.8950 0.8652 0.9061 0.8813 0.8869

Slide 18

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Moving Average Forecasts

The criticism against the simple moving average is that we use equal weights to denote all previous values. To Deal with this problem we use exponential smoothing where the forecast is the average of data from w periods prior to the forecast data point.
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Slide 19

Barometric Methods
National Bureau of Economic Research Department of Commerce Leading Indicators Lagging Indicators Coincident Indicators Composite Index Diffusion Index
Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 20

Prepared by Robert F. Brooker, Ph.D.

Econometric Models
Single Equation Model of the Demand For Cereal (Good X)
QX = a0 + a1PX + a2Y + a3N + a4PS + a5PC + a6A + e QX = Quantity of X PX = Price of Good X Y = Consumer Income N = Size of Population
Prepared by Robert F. Brooker, Ph.D.

PS = Price of Muffins PC = Price of Milk A = Advertising e = Random Error

Slide 21

Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.

Econometric Models
Multiple Equation Model of GNP
Ct ! a1  b1GNPt  u1t I t ! a2  b2T t 1  u2t GNPt | Ct  I t  Gt

Reduced Form Equation

a1  a2 b2T t 1 Gt GNPt !   b1  1  b1 1 1  b1
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 22

Input-Output Forecasting
The example of such forecasting is investment demand as a result of increase in GDP and GDP increase as a result of investment demand. In the other sense we can say it forecasting the demand of one thing as a result of change in some thing else.
Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved. Slide 23